BPC-157 vs TB-500 — What the Preclinical Evidence Actually Compares

Ask about BPC-157 vs TB-500 and you will usually get an answer that treats them as two versions of the same idea. They occupy the same shelf in almost every research peptide catalogue and are frequently sold as a blend. They are not similar molecules, they do not act through similar mechanisms, and the quality of evidence behind each is different in kind rather than degree.

What These Compounds Are

BPC-157 is a 15-amino-acid peptide, sequence GEPPPGKPADDAGLV, molecular weight 1419.5 Da. It was first described in 1993 by a group at the University of Zagreb School of Medicine, who reported isolating a roughly 40 kDa protein from human gastric juice — which they named BPC, for Body Protection Compound — and identified this 15-residue fragment as the essential active portion.

A point of accuracy that most supplier literature gets wrong: the parent protein has not, to our knowledge, been deposited in any public sequence database. The 1993 paper characterises it by apparent molecular weight alone. It is therefore not defensible to describe BPC-157 flatly as "a naturally occurring human peptide" — the honest formulation is that its discoverers described it as a partial sequence of a gastric juice protein that has never been independently sequenced.

Its composition is unusual and relevant to handling. BPC-157 contains no methionine, cysteine, tryptophan, asparagine or glutamine — removing the principal oxidative and deamidation degradation routes that dominate most peptides. Its main chemical liability is hydrolysis at the Asp10–Asp11 bond.

TB-500 is where the market gets genuinely confused, and the confusion has an analytical consequence. TB-500 is not thymosin beta-4. It is the N-terminally acetylated seven-residue actin-binding fragment, Ac-LKKTETQ, corresponding to residues 17–23 of Tβ4, with a molecular weight of 889.0 Da. This has been confirmed three independent ways: by high-resolution Orbitrap mass spectrometry performed on actual purchased TB-500 product, by equine doping control analysis, and by the FDA’s own nomenclature, which lists the substance as "Thymosin Beta-4, Fragment (LKKTETQ)."

Mechanism: A Defined Target Versus a Pleiotropic One

Thymosin beta-4 is the principal actin-monomer-sequestering protein of mammalian cells, established in the early 1990s, and the LKKTETQ motif is its actin-binding site. The landmark functional work is Bock-Marquette and colleagues in Nature (2004): Tβ4 forms a complex with PINCH and integrin-linked kinase, activating Akt; in mice after coronary artery ligation it improved early myocyte survival and cardiac function. Angiogenic activity in endothelial cells and anti-inflammatory activity through blockade of NF-κB p65 nuclear translocation are separately documented.

BPC-157 has no identified receptor. Its effects are described as pleiotropic modulation of several systems — and, a point worth making because it cuts against the usual criticism, the flagship mechanistic papers come from an independent group in Taiwan, not from Zagreb:

• VEGFR2–Akt–eNOS signalling (J Mol Med, 2017). Note carefully: BPC-157 upregulated VEGF receptor 2, not VEGF-A itself. This distinction is routinely mangled.
• FAK–paxillin phosphorylation in rat tendon explants and isolated fibroblasts (J Appl Physiol, 2011). The same study found it did not increase proliferation; it increased migration, spreading and survival under oxidative stress.
• Growth hormone receptor upregulation in cultured rat tendon fibroblasts (Molecules, 2014), with JAK2 activated only when growth hormone was also present — a permissive, sensitising effect. BPC-157 did not raise growth hormone.
• Src–caveolin-1–eNOS-mediated vasodilation in isolated rat aorta (Sci Rep, 2020).

Reported dopaminergic and serotonergic effects rest on rodent behavioural pharmacology, not receptor binding data. No published binding or occupancy study exists.

Evidence Quality: Two Different Problems

BPC-157’s problem is concentration. A PubMed query for "BPC 157" on 14 September 2026 returned 228 indexed records, of which 176 — just over 77% — carry Sikirić or Seiwerth as an author, and 177 carry a Zagreb affiliation. Zero are indexed as clinical trials or randomised controlled trials.

The nuance most coverage misses: this applies primarily to the efficacy literature. The mechanistic work and the one formal pharmacokinetic study come from independent groups. The accurate statement is that efficacy claims rest heavily on one laboratory while mechanistic and PK characterisation has been independently produced — not that a single lab did everything.

A 2026 review in Pharmaceutics adds sharper methodological criticism: virtually all preclinical studies use a single dose level, precluding any dose–response relationship; no study has characterised the relationship between plasma concentration and observed effect; and human exposure amounts to fewer than 30 subjects across three uncontrolled pilot studies using non-standardised preparations. A registered Phase I trial was terminated without publishing results.

Its pharmacokinetics were characterised independently in 2022 at Air Force Medical University, Xi’an: plasma half-life under 30 minutes in both rats and beagles, linear kinetics, intramuscular bioavailability of 14–19% in rats and 45–51% in dogs. That produces a genuine puzzle — a sub-30-minute half-life against reported effects lasting hours to days.

The widely repeated claim that BPC-157 is "stable in human gastric juice for more than 24 hours" deserves scepticism. Tracing the citation chain leads only to review articles by the same group citing one another. No primary experimental stability study with methods and data appears to exist, and the structural explanation usually offered — a polyproline II helix conferring protease resistance — has not been confirmed by circular dichroism, NMR or crystallography.

Thymosin beta-4’s problem is the opposite. It has been trialled extensively in humans — and has consistently failed on primary efficacy endpoints while proving consistently safe. ARISE-3 (dry eye, Phase 3, n=700) missed its primary endpoint. SEER-1 (neurotrophic keratopathy, Phase 3, n=18) terminated early at p = 0.0656. SEER-3 (EU, 2025) missed its primary endpoint. A Phase 2 pressure ulcer trial (n=72) met safety but not efficacy. A Phase 1 IV study in 40 healthy volunteers found no dose-limiting toxicities.

The scientific basis for using the fragment at all rests on a specific 2003 study in diabetic and 26-month-aged mice, which concluded that the seven-amino-acid synthetic peptide LKKTETQ was able to promote repair comparable to the parent molecule. That is a real finding — in aged mice, on dermal wound endpoints, using the non-acetylated fragment. A 2024 metabolism study complicates it considerably: using high-resolution mass spectrometry across human serum, enzyme systems and rat urine alongside fibroblast scratch assays, the authors found Ac-LK dominant as an early metabolite and reported that only Ac-LKKTE showed significant wound-healing activity — concluding that previously reported TB-500 activity may be due to its metabolite rather than the parent form.

The Head-to-Head Study

Until July 2026 no primary study had compared them directly. There is now exactly one. Published in Jt Dis Relat Surg, the design was 32 male Sprague-Dawley rats aged 12 weeks: standardised Achilles transection and repair, four arms of eight — control, BPC-157, TB-500, and the combination — dosed intraperitoneally daily and assessed at four weeks.

• Maximum load to failure was higher in both treatment arms, but reached significance only for TB-500 (p = 0.041; BPC-157 p = 0.389).
• Total Bonar score was significantly better for TB-500 only (p = 0.016).
• Total Movin score was significantly better for TB-500 (p = 0.017) and the combination (p = 0.040).
• Collagen type I expression showed no significant difference between any groups (p = 0.242). Collagen type III did differ significantly (p < 0.0001), elevated in the TB-500 arm — a shift in collagen subtype ratio rather than an increase in the mature load-bearing collagen.
• Combined BPC-157 and TB-500 treatment did not confer additional benefits compared to either agent alone.

How to read this responsibly matters. Eight animals per group, a single dose level for each compound, one four-week timepoint, described by its own authors as exploratory. It does not establish that TB-500 outperforms BPC-157 — the arms were not dose-matched or dose-optimised, so the comparison is between two arbitrary doses rather than two compounds.

The genuinely robust finding is the negative one: no additive effect from combination. The authors’ proposed explanation — convergence on shared downstream pathways — is explicitly labelled as requiring confirmation. For a market that sells the two as a blend, this is the most commercially inconvenient and best-evidenced result available.

Regulatory Position

Both are prohibited in sport at all times, but under different provisions, and getting these the wrong way round is a common tell. BPC-157 was added to the WADA Prohibited List for 2022 under S0, Non-Approved Substances — the category that exists because nothing else covers it. Thymosin-β4 and its derivatives, e.g. TB-500, has been named explicitly since the 2018 List under S2, Growth Factors. Both are non-specified substances, and neither holds a marketing authorisation in the UK, EU or US.

The FDA position has moved twice in 2026 and most published summaries are behind it. In April 2026, both compounds were removed from the agency’s higher-risk compounding category (503A Category 2) after their nominators withdrew, and now appear under bulk substances nominated but withdrawn. Withdrawal is not approval: the FDA’s stated concern remains that it lacks sufficient information to know whether the drug would cause harm when administered to humans, citing immunogenicity risk from aggregation and peptide-related impurities.

Then on 23–24 July 2026, the Pharmacy Compounding Advisory Committee voted 8–6 with one abstention in favour of adding both BPC-157 and TB-500 to the 503A Bulks List, against the opposition of FDA career scientists. The committee’s recommendation is non-binding, the FDA has not accepted it, and notice-and-comment rulemaking has not taken place. Neither compound is lawfully compoundable today — but anyone reading a summary written before July 2026 is missing the most significant development of the year in either direction.

Where This Leaves the Comparison

• Size: BPC-157 is 15 aa / 1419.5 Da; TB-500 is 7 aa / 889.0 Da.
• Molecular target: BPC-157 none identified; TB-500 binds G-actin via a characterised WH2 motif.
• Degradation liabilities: BPC-157 Asp10–Asp11 hydrolysis, no Met/Cys/Trp/Asn/Gln; TB-500 rapid exopeptidase trimming.
• Plasma half-life: BPC-157 under 30 min (rat, beagle); TB-500 not formally reported.
• Human trials: both zero controlled. Full-length Tβ4 reached Phase 3 — a different molecule.
• WADA: BPC-157 S0 since 2022; TB-500 S2 since 2018.

The defensible summary is not that one is better. It is that TB-500 has a defined molecular target and no human data; BPC-157 has no identified receptor and an efficacy literature concentrated in one institution. Those are different evidentiary problems, and neither is solved by combining the two.

Related Research